• "Artist's palette" model
• Start with blank canvas and build up from there
• Start with plot function (or similar)
• Use annotation functions to add/modify (text, lines, points, axis)

• Convenient, mirrors how we think of building plots and analyzing data
• Can't go back once plot has started (i.e. to adjust margins); need to plan in advance
• Difficult to 'translate' to others once a new plot has been created (no graphical 'language')
• Plot is just a series of R commands

• Plots are created with a single function call (xyplot, bwplot, etc.)
• Most useful for conditioning types of plots: Looking at how \(y\) changes with \(x\) across levels of \(z\)
• Things like margins/spacing set automatically because entire plot is specified at once
• Good for putting many many plots on a screen

• Sometimes awkward to specify an entire plot in a single function call
• Annotation in plot is not intuitive
• Use of panel functions and subscripts difficult to wield and requires intense preparation
• Cannot 'add' to the plot once it's created

• Split the difference between base and lattice
• Automatically deals with spacings, text, titles but also allows you to annotate by adding +
• Superficial similarity to lattice but generally easier/more intuitive to use
• Default mode makes many choices for you (but you can customize!)

• An implementation of The Grammar of Graphics by Leland Wilkinson
• Written by Hadley Wickham (while he was a graduate student at Iowa State)
• A 'third' graphics system for R (along with base and lattice)
• Web site

• Grammar of graphics represents an abstraction of graphics ideas/objects
• Think 'verb', 'noun', 'adjective' for graphics
• Allows for a 'theory' of graphics on which to build new graphics and graphics objects

"In brief, the grammar tells us that a statistical graphic is a mapping from data to aesthetic attributes (colour, shape, size) of geometric objects (points, lines, bars). The plot may also contain statistical transformations of the data and is drawn on a specific coordinate system"

• data frame
• aesthetic mappings: how data are mapped to color, size, thickness, …
• geoms: geometric objects like points, lines, shapes
• facets: for conditional plots (subgraph)
• stats: statistical transformations like binning, quantiles, smoothing.
• scales: what scale an aesthetic map uses (example: male = red, female = blue)
• coordinate system

• Works much like the plot function in base graphics system
• Looks for data in a data frame, similar to lattice
• Plots are made up of aesthetics (size, shape, color) and geoms (points, lines)

• Factors are important for indicating subsets of the data; they should be labeled
• The qplot() hides what goes on underneath, which is okay for most operations
• ggplot() is the core function and very flexible for doing things qplot() cannot do

library(ggplot2)
# qplot(x, y, data) -> scatter plot
qplot(carat, price, data=diamonds)

qplot(carat, price, data=diamonds, color=clarity)

# add smooth curve
qplot(carat, price, data=diamonds, 
      geom=c("point", "smooth")) 

qplot(price, data=diamonds, fill=cut)

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

# facets = . ~ cut (horizontal)
qplot(carat, price, data=diamonds, 
      facets=.~cut)

# facets = cut ~ . (vertical)
qplot(carat, price, data=diamonds, 
      facets=cut~.)

qplot(hwy, data = mpg, facets = drv ~ ., binwidth = 2)

• The qplot() function is the analog to plot() but with many built-in features
• Syntax somewhere in between base/lattice
• Produces very nice graphics, essentially publication ready (if you like the design)
• Difficult to go against the grain/customize (don't bother; use full ggplot2 power in that case)

• When building plots in ggplot2 (rather than using qplot) the "artist’s palette" model may be the closest analogy
• Plots are built up in layers
• Plot the data
• Overlay a summary
• Metadata and annotation

• Labels: xlab(), ylab(), labs(), ggtitle()
• Each of the geom functions has options to modify
• For things that only make sense globally, use theme()
• Example: theme(legend.position = "none")
• Two standard appearance themes are included
• theme_gray(): The default theme (gray background)
• theme_bw(): More stark/plain

• ggthemes packages Website
• xkcd packages Website

# aes: Aesthetic attributes, color, shape, size, thickness
# geom_*: Geometric objects, point, line, boxplot, bar
ggplot(diamonds, aes(x = carat, y = price)) + 
  geom_point()

# aes: Aesthetic attributes, color, shape, size, thickness
# geom_*: Geometric objects, point, line, boxplot, bar
ggplot(diamonds, aes(x = cut, y = price)) + 
  geom_boxplot()

# facet_grid: add subgraph
ggplot(diamonds, aes(x = carat, y = price)) + 
  geom_point() +
  facet_grid(cut~.)

# facet_grid: add subgraph
ggplot(diamonds, aes(x = carat, y = price)) + 
  geom_point() +
  facet_grid(.~cut)

# geom_smooth: method='lm' (linear regression)
ggplot(diamonds, aes(x = carat, y = price)) + 
  geom_point() +
  facet_grid(cut~.) + 
  geom_smooth(method='lm')

# color = cut
ggplot(diamonds, aes(x = carat, y = price, color = cut)) + 
  geom_point() +
  geom_smooth(method='lm')

testdat <- data.frame(x = 1:100, y = rnorm(100))
testdat[50, 2] <- 100  # Outlier!
plot(testdat$x, testdat$y, type = "l", ylim = c(-3,3))

g <- ggplot(testdat, aes(x = x, y = y))
g + geom_line()

g + geom_line() + ylim(-3, 3)

g + geom_line() + coord_cartesian(ylim = c(-3, 3))

• Standard deviation (SD)
• Standard error (SE)
• Confidence interval (CI)

library(datasets)
airquality$Month <- as.factor(airquality$Month)
airquality.mean <- aggregate(Ozone~Month, airquality, mean)
ggplot() + 
  geom_bar(data=airquality.mean, aes(x=Month, y=Ozone), stat="identity")

airquality.sd <- aggregate(Ozone~Month, airquality, sd)
airquality.eb <- merge(airquality.mean, airquality.sd, by="Month")
ggplot(data=airquality.eb) +
  geom_bar(aes(x=Month, y=Ozone.x), stat = "identity")+
  geom_errorbar( # Ozone.x = mean, Ozone.y = sd
    aes(x=Month, ymin=Ozone.x-Ozone.y, ymax=Ozone.x+Ozone.y), width=.1)

• The ggplot2 book by Hadley Wickham
• The R Graphics Cookbook by Winston Chang (examples in base plots and in ggplot2)
• ggplot2 web site
• ggplot2 doc
• ggplot cheat sheet
• r blooger

Reference

nba <- read.csv("http://datasets.flowingdata.com/ppg2008.csv")
head(nba)

##             Name  G  MIN  PTS  FGM  FGA   FGP FTM FTA   FTP X3PM X3PA
## 1   Dwyane Wade  79 38.6 30.2 10.8 22.0 0.491 7.5 9.8 0.765  1.1  3.5
## 2  LeBron James  81 37.7 28.4  9.7 19.9 0.489 7.3 9.4 0.780  1.6  4.7
## 3   Kobe Bryant  82 36.2 26.8  9.8 20.9 0.467 5.9 6.9 0.856  1.4  4.1
## 4 Dirk Nowitzki  81 37.7 25.9  9.6 20.0 0.479 6.0 6.7 0.890  0.8  2.1
## 5 Danny Granger  67 36.2 25.8  8.5 19.1 0.447 6.0 6.9 0.878  2.7  6.7
## 6  Kevin Durant  74 39.0 25.3  8.9 18.8 0.476 6.1 7.1 0.863  1.3  3.1
##    X3PP ORB DRB TRB AST STL BLK  TO  PF
## 1 0.317 1.1 3.9 5.0 7.5 2.2 1.3 3.4 2.3
## 2 0.344 1.3 6.3 7.6 7.2 1.7 1.1 3.0 1.7
## 3 0.351 1.1 4.1 5.2 4.9 1.5 0.5 2.6 2.3
## 4 0.359 1.1 7.3 8.4 2.4 0.8 0.8 1.9 2.2
## 5 0.404 0.7 4.4 5.1 2.7 1.0 1.4 2.5 3.1
## 6 0.422 1.0 5.5 6.5 2.8 1.3 0.7 3.0 1.8

library(reshape2) # for melt()
nba.m <- melt(nba, id.vars = "Name") # wide to long based on name
head(nba.m, 10)

##                Name variable value
## 1      Dwyane Wade         G    79
## 2     LeBron James         G    81
## 3      Kobe Bryant         G    82
## 4    Dirk Nowitzki         G    81
## 5    Danny Granger         G    67
## 6     Kevin Durant         G    74
## 7     Kevin Martin         G    51
## 8     Al Jefferson         G    50
## 9       Chris Paul         G    78
## 10 Carmelo Anthony         G    66

library(ggplot2) # for ggplot()
ggplot(nba.m, aes(variable, Name)) +
    geom_tile(aes(fill = value), colour = "white") +
    scale_fill_gradient(low = "white", high = "steelblue")

head(nba, 2)

##            Name  G  MIN  PTS  FGM  FGA   FGP FTM FTA   FTP X3PM X3PA  X3PP
## 1  Dwyane Wade  79 38.6 30.2 10.8 22.0 0.491 7.5 9.8 0.765  1.1  3.5 0.317
## 2 LeBron James  81 37.7 28.4  9.7 19.9 0.489 7.3 9.4 0.780  1.6  4.7 0.344
##   ORB DRB TRB AST STL BLK  TO  PF
## 1 1.1 3.9 5.0 7.5 2.2 1.3 3.4 2.3
## 2 1.3 6.3 7.6 7.2 1.7 1.1 3.0 1.7

nba[, 2:21] <- apply(nba[, 2:21], 2, scale) # scale, average to 0
head(nba, 2)

##            Name         G       MIN      PTS      FGM      FGA       FGP
## 1  Dwyane Wade  0.6179300 1.0019702 3.179941 2.920022 2.596832 0.5136017
## 2 LeBron James  0.7693834 0.6119299 2.566974 1.957185 1.697237 0.4649190
##        FTM      FTA        FTP       X3PM      X3PA        X3PP
## 1 1.917475 2.110772 -0.7401673 -0.1080044 0.1303647 -0.15749098
## 2 1.778729 1.896589 -0.5233214  0.4920201 0.6971679  0.02738974
##           ORB        DRB        TRB      AST      STL       BLK       TO
## 1 -0.27213551 -0.3465676 -0.3287465 1.652247 2.558238 1.2064646 1.790445
## 2 -0.06117775  1.0080940  0.6605370 1.516147 1.367252 0.8627425 1.059651
##           PF
## 1 -0.2984568
## 2 -1.3903719

nba.m <- melt(nba)
ggplot(nba.m, aes(variable, Name)) +
    geom_tile(aes(fill = value),colour = "white") +
    scale_fill_gradient(low = "white",high = "steelblue")

• Use treemapify packages
• Reference
• G20 Dataset

if (!require('treemapify')){
    library(devtools)
    install_github("wilkox/treemapify")
    library(treemapify)
}
data(G20)
head(G20)

##          Region        Country Trade.mil.USD Nom.GDP.mil.USD   HDI
## 1        Africa   South Africa        208000          384315 0.629
## 2 North America  United States       3969000        15684750 0.937
## 3 North America         Canada        962600         1819081 0.911
## 4 North America         Mexico        756800         1177116 0.775
## 5 South America         Brazil        494800         2395968 0.730
## 6 South America      Argentina        152690          474954 0.811
##   Population Economic.classification
## 1   53000000              Developing
## 2  316173000                Advanced
## 3   34088000                Advanced
## 4  112211789              Developing
## 5  201032714              Developing
## 6   40117096              Developing

# treemapify: convert format
# area, fill, label, group
treeMapCoordinates <- treemapify(data=G20,
    area = "Nom.GDP.mil.USD", fill = "HDI",
    label = "Country", group = "Region")
head(treeMapCoordinates)

##    fill           label     xmin     xmax     ymin      ymax         group
## 1 0.876  European Union  0.00000 38.66972  0.00000  58.99641        Europe
## 2 0.920         Germany 38.66972 63.32079  0.00000  19.17284        Europe
## 3 0.893          France 38.66972 63.32079 19.17284  33.88097        Europe
## 4 0.875  United Kingdom 38.66972 63.32079 33.88097  47.64081        Europe
## 5 0.881           Italy 38.66972 63.32079 47.64081  58.99641        Europe
## 6 0.937   United States  0.00000 53.16491 58.99641 100.00000 North America

ggplotify(treeMapCoordinates)

ggplotify(treeMapCoordinates)+ 
    scale_fill_gradient(low = "white",high = "steelblue")

• Choropleth map?
• Stat data to map
• choroplethr & choroplethrMaps packages

• rgdal, rgeos,maptools packages for map files
• ggplot2 & RColorBrewer for plotting

## [1] TRUE

• Use Taiwan open data
• Include longitude, latitute, boundary, …
• shapefile Wiki
• Spatial data format
• Reference

• Use maptools package, readShapeSpatial function

.packages <- c("rgdal", "rgeos", "maptools", "ggplot2", "gpclib", "RColorBrewer")
.inst <- .packages %in% installed.packages()
if(length(.packages[!.inst]) > 0) suppressMessages(install.packages(.packages[!.inst], repos="http://cran.rstudio.com/"))
invisible(lapply(.packages, library, character.only=TRUE))

tw_new <- readShapeSpatial("Taiwan/Town_MOI_1041215.shp")
names(tw_new)

##  [1] "OBJECTID"   "T_UID"      "Town_ID"    "T_Name"     "T_Desc"    
##  [6] "Add_Date"   "Add_Accept" "Remark"     "County_ID"  "C_Name"

• rgdal, rgeos,maptools
• fortify: shapefile \(\rightarrow\) data.frame
• Reference

head(tw_new$Town_ID)

## [1] 1001402 1001321 1000913 1001411 1001416 1000712
## 368 Levels: 0900701 0900702 0900703 0900704 0902001 0902002 ... 6801300

tw_new.df <- fortify(tw_new, region = "T_UID") # from ggplot2 package

• Reference

head(tw_new.df,10)

##        long      lat order  hole piece id group
## 1  119.9170 26.17518     1 FALSE     1  1   1.1
## 2  119.9171 26.17517     2 FALSE     1  1   1.1
## 3  119.9171 26.17518     3 FALSE     1  1   1.1
## 4  119.9171 26.17518     4 FALSE     1  1   1.1
## 5  119.9171 26.17518     5 FALSE     1  1   1.1
## 6  119.9172 26.17518     6 FALSE     1  1   1.1
## 7  119.9172 26.17518     7 FALSE     1  1   1.1
## 8  119.9172 26.17518     8 FALSE     1  1   1.1
## 9  119.9173 26.17515     9 FALSE     1  1   1.1
## 10 119.9173 26.17515    10 FALSE     1  1   1.1

# Randomly generate some data for plotting
mydata <- data.frame(NAME_2=tw_new$T_Name, id=tw_new$T_UID,
                     prevalence=rnorm(length(tw_new$T_UID)))
head(mydata)

##                   NAME_2  id prevalence
## 1 \xa6\xa8\xa5\\\xc2\xed 178  0.5066914
## 2            \xa8ΥV\xb6m 164  1.8918956
## 3     \xb3\xc1\xbcd\xb6m 118  0.8782975
## 4     \xba\xf1\xaeq\xb6m 376 -0.9562235
## 5  \xc4\xf5\xc0\xac\xb6m 369  0.8223115
## 6      \xa5Ф\xa4\xc2\xed  78 -0.4283949

# use iconv to convert
# from big5 to utf-8
mydata$NAME_2 <- iconv(as.character(mydata$NAME_2),
                       from="big5", to="UTF-8")
head(mydata,10)

##    NAME_2  id prevalence
## 1  成功鎮 178  0.5066914
## 2  佳冬鄉 164  1.8918956
## 3  麥寮鄉 118  0.8782975
## 4  綠島鄉 376 -0.9562235
## 5  蘭嶼鄉 369  0.8223115
## 6  田中鎮  78 -0.4283949
## 7  社頭鄉  83 -1.0360082
## 8  竹田鄉 157  0.5036374
## 9  萬丹鄉 148  0.1431512
## 10 三灣鄉  64 -0.8861121

final.plot <- merge(tw_new.df, mydata,by="id", all.x=T)
head(final.plot,10)

##    id     long      lat order  hole piece group NAME_2 prevalence
## 1   1 119.9170 26.17518     1 FALSE     1   1.1 南竿鄉   1.473039
## 2   1 119.9171 26.17517     2 FALSE     1   1.1 南竿鄉   1.473039
## 3   1 119.9171 26.17518     3 FALSE     1   1.1 南竿鄉   1.473039
## 4   1 119.9171 26.17518     4 FALSE     1   1.1 南竿鄉   1.473039
## 5   1 119.9171 26.17518     5 FALSE     1   1.1 南竿鄉   1.473039
## 6   1 119.9172 26.17518     6 FALSE     1   1.1 南竿鄉   1.473039
## 7   1 119.9172 26.17518     7 FALSE     1   1.1 南竿鄉   1.473039
## 8   1 119.9172 26.17518     8 FALSE     1   1.1 南竿鄉   1.473039
## 9   1 119.9173 26.17515     9 FALSE     1   1.1 南竿鄉   1.473039
## 10  1 119.9173 26.17515    10 FALSE     1   1.1 南竿鄉   1.473039

library(RColorBrewer)
twcmap <- ggplot() +
    geom_polygon(data = final.plot, 
                 aes(x = long, y = lat, group = group, 
                     fill = prevalence), 
                 color = "black", size = 0.25) + 
    coord_map()+
    scale_fill_gradientn(colours = brewer.pal(9,"Reds"))+
    theme_void()+
    labs(title="Prevalence of X in Taiwan")

twcmap

• ggmap package

if (!require('ggmap')){
    install.packages("ggmap")
    library(ggmap)
}
twmap <- get_map(location='Taiwan', zoom=7, language="en-US")
# location: name or long/lat
# zoom: 2-20
# language: map language

ggmap(twmap)

# maptype: "terrain", "terrain-background", "satellite",
# "roadmap","hybrid" (google maps), "terrain", "watercolor", 
# "toner" (stamen maps), 
# or a positive integer for cloudmade maps (see ?get_cloudmademap)
tpmap <- get_map(location = c(121.43, 24.93, 121.62, 25.19), 
                     zoom = 14, maptype = 'roadmap')
ggmap(tpmap, extent = 'device') # extent='device': full screen

ggmap(twmap) + # ggmap
    geom_polygon(data = final.plot,
        aes(x = long, y = lat, group = group, fill = prevalence), 
        color = "grey80", size = 0.1,alpha = 0.5) + 
  scale_fill_gradientn(colours = brewer.pal(9,"Reds"))

# Get location of each state
StateCenter <- data.frame( 
    region=tolower(state.name),
    lon=state.center$x,
    lat=state.center$y)
head(StateCenter, 1)

##    region      lon     lat
## 1 alabama -86.7509 32.5901

# Population data
StatePop <- merge(df_pop_state, StateCenter, by="region") 
head(StatePop, 1)

##    region   value      lon     lat
## 1 alabama 4777326 -86.7509 32.5901

# Value to point
PopPoint <- NULL 
for(i in 1:nrow(StatePop)){
    # 1M people -> one point
    for(j in 1:(StatePop[i, "value"] / 1000000)){
        PopPoint <- rbind(PopPoint, StatePop[i, ])   
    }
}
head(PopPoint, 3)

##    region   value      lon     lat
## 1 alabama 4777326 -86.7509 32.5901
## 2 alabama 4777326 -86.7509 32.5901
## 3 alabama 4777326 -86.7509 32.5901

USMap <- get_map(location = "United States", zoom = 4)
densityMap <- ggmap(USMap, extent = "device") + 
  geom_density2d(data = PopPoint, aes(x = lon, y = lat), size = 0.3) +
  stat_density2d(data = PopPoint, 
                 aes(x = lon, y = lat, fill = ..level.., alpha = ..level..), 
                 size = 0.01, bins = 16, geom = "polygon") + 
  scale_fill_gradient(low = "green", high = "red", guide = FALSE) + 
  scale_alpha(range = c(0, 0.3), guide = FALSE)

densityMap

• ggmap package source code
• ggmap cheat sheet
• ggmap doc

• Data ggplot(Data)
• Geometries geom_line(), geom_bar(), geom_point()
• Aesthetic properties ggplot(Data, aes(x=1, y=1, color='black'))
• Scale mappings scale_fill_continuous(), scale_fill_grey()
• Stats stat_density2d()

• ggplot2 Cheat Sheet
• Courtesy by Professor Yi-Ju Tseng, Chang Gung University, Taiwan